Novel 2,7-Diazaspiro[4,4]nonane Derivatives to Inhibit Mouse and Human Osteoclast Activities and Prevent Bone Loss in Ovariectomized Mice without Affecting Bone Formation Lucile Mounier, Anne Morel, Yann Ferrandez, Jukka Morko, Jukka Vääräniemi, Marine Gilardone, Didier Roche, Jacqueline Cherfils, Anne Blangy To cite this version: Lucile Mounier, Anne Morel, Yann Ferrandez, Jukka Morko, Jukka Vääräniemi, et al.. Novel 2,7- Diazaspiro[4,4]nonane Derivatives to Inhibit Mouse and Human Osteoclast Activities and Prevent Bone Loss in Ovariectomized Mice without Affecting Bone Formation. Journal of Medicinal Chemistry, American Chemical Society, 2020, 10.1021/acs.jmedchem.0c01201. hal-03006787 HAL Id: hal-03006787 https://hal.archives-ouvertes.fr/hal-03006787 Submitted on 16 Nov 2020 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Novel 2,7-Diazaspiro[4,4]nonane derivatives to inhibit mouse and human osteoclast activities and to prevent bone loss in ovariectomized mice without affecting bone formation. Lucile Mounier†, Anne Morel†, Yann Ferrandez‡, Jukka Morko∥, Jukka Vääräniemi∥, Marine Gilardone⊥, Didier Roche⊥, Jacqueline Cherfils‡, Anne Blangy†,* † Centre de Recherche de Biologie Cellulaire de Montpellier, CRBM, Univ Montpellier, CNRS, 34000 Montpellier, France. ‡ Laboratoire de Biologie et Pharmacologie Appliquée, CNRS, Ecole Normale Supérieure Paris-Saclay and Université Paris-Saclay, 91190 Gif-sur-Yvette, France ∥ Pharmatest Services Ltd., Itäinen Pitkäkatu 4, 20520 Turku, Finland ⊥ Edelris, 60 avenue Rockefeller, 69008 Lyon, France. *Corresponding author: BLANGY Anne, [email protected] Postal address: CNRS UMR 5237 CRBM 1919 Route de Mende - 34293 Montpellier Cedex 5- FRANCE Tel : +33 434 359 508 - Fax : +33 434 359 410 1 ABSTRACT Osteoporosis is currently treated with drugs targeting the differentiation or viability osteoclasts, the cells responsible for physiological and pathological bone resorption. Nevertheless, osteoporosis drugs that would target only osteoclast activity are expected to preserve bone formation by osteoblasts, in contrast with current treatments. We report here the design, synthesis and biological characterization of a series of novel N-arylsufonamides featuring a diazaspiro[4,4]nonane nucleus to target the guanine nucleotide exchange activity of DOCK5, which is essential for bone resorption by osteoclasts. These compounds can inhibit both mouse and human osteoclast activity. In particular, 4-chlorobenzyl-4-hydroxy-2- phenyl-1-thia-2,7-diazaspiro[4,4]nonane 1,1-dioxide (compound E197) prevented pathological bone loss in the mouse. Most interestingly, treatment E197 did not affect osteoclast and osteoblast numbers and hence did not impair bone formation. E197 could represent a lead molecule to develop new anti-osteoporotic drugs targeting the mechanism of osteoclast adhesion onto the bone. KEY WORDS Osteoporosis, Preclinical Studies, Osteoclast, diazaspiro[4,4]nonane, Bone histomorphometry 2 INTRODUCTION Bone is a dynamic tissue undergoing permanent remodeling throughout life: it is continually resorbed by osteoclasts, which are myeloid cells, and replaced by osteoblasts, which are mesenchymal cells1,2. Bone modeling and remodeling are necessary for fracture healing and adaptation of the skeleton to load and to mechanical use, as well as for calcium and phosphorus homeostasis. An imbalance between bone resorption and bone formation results in several bone diseases. In particular, excessive resorption activity of osteoclasts, resulting in bone loss and eventually osteoporosis, occurs in a number of physiopathological conditions. Osteoclast hyperactivity indeed accompanies hormone deficiency after menopause 3, rheumatoid arthritis and other inflammatory diseases4,5, multiple myeloma and numerous cancers that metastasize to the bone 6, corticotherapy7, and also the lack of mechanical forces applied to the bone, as for instance upon prolonged bed rest or spinal cord injury8. Several infectious diseases are also characterized by abnormal bone loss9. Currently, the most widely used clinical treatments to prevent osteoporosis target osteoclast differentiation and survival. The bisphosphonates are pyrophosphate analogs that bind to the bone, they are ingested by the bone-resorbing osteoclasts and cause their apoptosis10, whereas monoclonal antibody denosumab targets receptor activator of nuclear factor-κB ligand (RANKL), the cytokine essential for osteoclast differentiation11. But these treatments hamper the stimulatory activity of osteoclasts on osteoblast differentiation and on bone formation activity; as a consequence, patients receiving these drugs suffer from a blockade of de novo bone formation12,13. In fact, osteoclasts were found able to stimulate bone formation, independent of bone resorption14. Whereas the mechanisms are not completely understood, several proteins secreted by non- resorbing osteoclasts and collectively called clastokines were found to stimulate osteoblast differentiation12, such as Wnt10b or the chemokine sphingosine-1-phosphate S1P15. Thus, targeting specifically osteoclast bone-resorption function, instead of osteoclast differentiation 3 and viability, appears an attractive strategy to control bone resorption without affecting bone formation14. Indeed, Odanacatib, an inhibitor of the key osteoclast protease Cathespin K, proved efficient at preventing pathological bone loss while preserving bone formation in patients but unfortunately, the molecule recently failed in clinical phase III trials due to increased risk of stroke16. Targeting osteoclast adhesion structures on bone also appears as a tempting strategy17, still in the preclinical or early clinical stage. The resorbing osteoclasts is attached on the bone through a unique adhesion structure based on a belt of podosomes; the podosomes attach osteoclasts on the bone matrix via αvβ3, the major osteoclast integrin and perturbing the organization of podosomes in osteoclasts prevents bone resorption17. Indeed, an inhibitor of αvβ3 allowed reducing osteoclast activity and proved efficient to protect women from postmenopausal bone loss in a phase I study18. Another approach is to aim at the regulators of the dynamics of actin, the major component of podosomes. We reported previously the identification of a small chemical compound inhibiting the activity of DOCK5, a guanine nucleotide exchange factor (GEF) of the small GTPase Rac essential to assemble the podosome belt in mouse osteoclasts19,20. The commercial molecule N-(3,5-Dichlorophenyl) benzenesulfonamide (C21, CAS 54129-15-6) disorganizes the podosome belt and hinders osteoclast activity20. When administered to the mouse, C21 prevented pathological bone loss, while bone formation by osteoblasts was maintained, making this a promising strategy to prevent pathological bone loss21. In this study, based on the structure of C21, we identified a identified a new N-arylsufonamides family of compounds featuring a [4,4]nonane nucleus that inhibit DOCK5 exchange activity. We showed that these compounds not only disorganized reversibly mouse osteoclast podosomes and prevented their activity, as C21 does20,21, but also hindered the activity of human osteoclasts, contrarily to C21. Moreover, we tested one of the N-arylsufonamide compounds in vivo in the mouse model of ovariectomy 4 (OVX)-induced bone loss and we found that could prevent pathological bone, while bone formation was preserved. 5 RESULTS AND DISCUSSION Identification of an inhibitor of human DOCK5 exchange activity. The inhibition of DOCK5 exchange activity towards the GTPase Rac with N-(3,5-Dichlorophenyl) benzenesulfonamide (C21, Figure 1A), prevents bone resorption by mouse osteoclasts by disorganizing their cytoskeleton and it efficiently protected mice against pathological bone loss in various mouse models of pathological bone loss without affecting bone formation20,21. This established that DOCK5 inhibition could represent an interesting therapeutic strategy against osteoporosis. Nevertheless, we found that C21 did not inhibit the activity of human peripheral blood cell-derived osteoclasts when used up to 30 µM (Supplementary Figure 1A). C21 was toxic for human osteoclasts above this dose (Supplementary Figure 1B), consistent with what we had observed previously on mouse osteoclasts20. Thus, C21 was not appropriate for further development toward clinical applications. 6 Figure 1: Chemical structures of the compounds. (A) C21: N-(3,5-Dichlorophenyl) benzenesulfonamide and E73: rac-(4S,5R)-7-(3-chlorobenzyl)-4-hydroxy-2-phenyl-1-thia-2,7-diazaspiro[4,4]nonane 1,1-dioxide. (B) General structure of E73 and its derivatives. (C-D) Structures of E73 analogs. E196: rac-(4S,5R)-7-(2- chlorobenzyl)-4-hydroxy-2-phenyl-1-thia-2,7-diazaspiro[4,4]nonane 1,1-dioxide, E197: rac-(4S,5R)-7-(4- chlorobenzyl)-4-hydroxy-2-phenyl-1-thia-2,7-diazaspiro[4,4]nonane 1,1-dioxide, E202: rac-(4S,5R)-7-(3,4- dichlorobenzyl)-4-hydroxy-2-phenyl-1-thia-2,7-diazaspiro[4,4]nonane 1,1-dioxide, E203: rac-(4S,5R)-4- hydroxy-2-phenyl-7-(quinolin-3-ylmethyl)-1-thia-2,7-diazaspiro[4,4]nonane